Sight Unseen
Mike May was blind most of his life until surgery gave him his sight back. But two years later he still can't recognize his own wife. By learning why, psychologists are revealing the very origins of vision
By Michael Abrams
Photography by Alyson Aliano
DISCOVER Vol. 23 No. 06 | June 2002

Mike May holds the world speed record for downhill skiing by a blind person. In his competitive days he would slalom down the steepest black-diamond slopes at 65 miles an hour, with a guide 10 feet ahead to shout "left" and "right." The directions were just obvious cues. The rest came from the feel of the wind racing against his cheeks and the sound of the guide's skis snicking over the snow. But May's days as a world-class blind athlete are behind him. He's no longer blind.
May lost his vision at the age of 3, when a jar of fuel for a miner's lantern exploded in his face. It destroyed his left eye and scarred the cornea of his right, but over the next 43 years he never let those disabilities slow him down. He played flag football in elementary school, soccer in college, and nearly any activity that didn't involve projectiles as an adult. He earned a master's degree in international affairs from Johns Hopkins, took a job with the CIA, and became the president and CEO of the Sendero Group, a company that makes talking Global Positioning Systems for the blind. Along the way, he found time to help develop the first laser turntable, marry, have two children, and buy a house in Davis, California. "Someone once asked me if I could have vision or fly to the moon, what would I choose," he once wrote. "No question— I would fly to the moon. Lots of people have sight, few have gone to the moon."
Then one November day in 1999, he came back to his senses. At St. Mary's Hospital in San Francisco, surgeon Daniel Goodman dropped a doughnut of corneal stem cells onto May's right eye (his left was too severely damaged to be repaired). The cells replaced scar tissue and rebuilt the ocular surface, preparing the eye for a corneal transplant. On March 7, 2000, when the wraps were removed, May got his first look at his wife, his children, and for the first time since he was a toddler, himself.
Sight restoration is a periodic miracle— both for its recipients and for the scientists who have the privilege of studying them. As early as the fifth century B.C., Egyptian surgeons used a needle to push their patients' cataract-covered lenses away from their pupils, affording them some degree of sight. More recently, in the late 1960s, surgeons learned to remove cataracts with ultrasound. The stem-cell surgery performed on May was developed in Japan and introduced in 1999. Since then hundreds of people have benefited from it. But of all those who have had their sight restored throughout history, only about 20 recorded cases were blind since childhood, and of those, most had less-than-perfect corneas after surgery. When Goodman peered into May's eye after the surgery, he saw a lens that ought to provide crystal-clear vision.
It doesn't— far from it. Pristine as his optical hardware is, May's brain has never been programmed to process the visual information it receives. May still travels with his dog, Josh, or taps the sidewalk with a cane, and refers to himself as "a blind man with vision." And that paradox fascinates Don MacLeod and Ione Fine, experimental psychologists at the University of California at San Diego. The speed with which babies learn to understand the world suggests that they're born with the ability to process some aspects of vision. But which aspects, exactly? What is learned and what is hardwired? During the past year and a half, Fine and MacLeod have put May through a battery of physical and psychological tests, including functional magnetic resonance imaging, or fMRI, which tracks blood flow in the brain. The results are opening the first clear view into how we learn to see.

Functional magnetic resonance imaging, here being performed on graduate student Melissa Sáenz, tracks blood flow in the brain. UCSD researchers used this same technique at Stanford University, in collaboration with the Salk Institute, to chart Mike May's visual processing after his sight was restored.

MacLeod's laboratory at the university is a labyrinth of filing cabinets, optical equipment, and oddly placed desks. "It's well booby-trapped," he says, steering May toward the first of many tests one afternoon. "But May has an uncanny ability to navigate complicated arrangements." Tall and athletic, with features that look boyishly handsome despite his graying black hair, May would make a good James Bond if not for a few side effects of his blindness. Unlike the rest of his body, his eyelids haven't had a lifelong workout. Perpetually half closed, they lend a stoic blankness to his face that's relieved only by the occasional smile. He has yet to learn facial expressions.
Sitting obligingly in front of an ancient computer monitor, May watches as thick black-and-orange bars appear on the screen. MacLeod and Fine are testing his ability to see detail. His job is to adjust the contrast with a trackball until he can just see the bars. A click on a mouse brings up another set of bars, thinner than the last, and he plays around with those until he can see them too. Although his right eye ought to provide 20/20 vision, in reality it's closer to 20/500. Instead of discerning the letter E on an eye chart from 25 feet, May can see it only from two. In the past the blurred vision of people with restored sight was blamed on scar tissue from surgery. But stem-cell surgery leaves no scars. The signals are reaching May's brain, but they are not being interpreted very well.
More than 300 years ago, in a famous letter to the philosopher John Locke, the Irish thinker William Molyneux anticipated what May sees. A blind man who is suddenly given vision, Molyneux suggested, wouldn't be able to tell the difference between a cube and a sphere. Sight is one kind of perception and touch another; they can be linked only through experience.
The most dramatic proof of this theory came in an experiment published in 1963 by Richard Held and Alan Hein, who were then professors at BrandeisUniversity in Waltham, Massachusetts. Held and Hein raised two kittens in total darkness. But every so often they would place the kittens in separate baskets, suspend the baskets from a single circular track, and turn on the lights. Both baskets hung just above the floor, but one had holes for the kitten's legs to poke through; the other did not. The free-limbed cat ran in circles on the floor, pulling the other basket along behind it; the other kitten had no choice but to sit and watch. While the active kitten learned to see normally, the passive kitten stayed effectively blind: Its eyes could see, but its brain never learned to interpret the sensory input.
Held and Hein's experiment has never been duplicated. But in the past half century, studies of sight restoration, most notably by Oliver Sacks and Richard Gregory, have verified that some things can't be understood without experience. Objects, faces, depth— just about everything that helps us function in the world— are meaningless when a person who has never seen before gets sight. "Babies are born into a bright, buzzing confusion, but we can't ask them what it's like," Fine says. "In some ways talking to Mike May is like getting to talk to a 7-month-old."

Ione Fine and Don MacLeod use this interferometer to shine a split laser beam into May's eyes to test his ability to process visual information. They also run the same test on themselves. "I spend a lot of my time down here," Fine says, "with an eye patch on and a bite bar in my mouth. It's disgusting . . . really old plastic."

In the first months after his surgery, May fulfilled Molyneux's prediction: He couldn't distinguish a sphere from a cube. Since then his sight has improved, but only slightly. He has a better grasp of spheres and squares ("We've shown him an awful lot of them," Fine says), and with practice he can understand things he's seen again and again. But this is only a work-around: He's past the critical period for learning to recognize objects instantly.
"Two of the major clues I have are color and context," May says. "When I see an orange thing on a basketball court, I assume it's round. But I may not be really seeing the roundness of it." Faces give him even more trouble. Although he has seen faces everywhere since the first day his vision was restored, they simply don't coalesce into recognizable people. Their expressions— their moods and personalities— elude him entirely. Even his wife is familiar to him only by the quality of her gait, the length of her hair, and the clothes she wears. "If a face has no hair and a fake moustache, we can still tell the gender," Fine says. "But he can't deal with it. The bit of the brain that does that isn't working."
The best proof of this can be seen in the basement, where MacLeod's interferometer sits. Designed to test the brain's ability to process visual information, the machine works by shining a split laser beam into a subject's eye. As the beams travel, their light waves interfere with each other, bypassing the optics of the cornea and projecting a pattern onto the retina. Most subjects who sit in front of the interferometer will see light and dark stripes, regardless of the quality of their optics. But when May opens his eyes to receive the beams, he sees nothing at all.
The interferometer results are backed by fMRI scans, which track May's brain activity as it's occurring. The scans show that when May sees faces and objects, the part of his brain that should be used to recognize them is inactive. But there's a catch. When he sees an object in motion, the motion-detection part of his brain lights up like a disco ball. He can interpret movement on a computer screen as well as any normal-sighted adult and seems to have the same skill in real life. "We were driving along, and a minivan came up to us pretty fast on his side," Fine remembers. "It whizzed by him, and he mentioned that it was going fast. That's a complicated calculation. The motion on the retina depends on how big the car is, how close, and how fast it's going."
It's hard to escape the conclusion that motion detection, unlike every other visual experience aside from color, is largely hardwired. The best illustration of this may be offered, once again, by cats. "If you roll a ball along a floor, the cat will chase it as long as it's moving," Fine says. "As soon as it's stationary, the cat will have a hard time seeing it and will ignore it." That's why mice freeze when they're afraid. It may also explain why May, who can barely recognize a stationary ball, is pretty good at catching a moving one. It's his favorite use of his new sense. "I don't know who has more fun," he says, "my 8-year-old or me."

When May walks down the street, he can't recognize perspective lines, so he uses visual landmarks to keep his bearings. "I'm learning one frame at a time," he says.

Blind people spend their entire lives understanding the world through their hands. Their memories, their mental maps of the places they know, their understanding of Labradors, doorknobs, and the moguls on a ski slope are all tactile. The sudden introduction of a new sense can't alter that fundamental way of experiencing the universe. Instead, any new information gleaned from light is simply graphed onto the original, tactile map. "The old idea that there is one picture of the world on the surface of the cortex is way too simple," MacLeod says. "In fact, we have a couple dozen complete maps." For someone just learning how to merge all that information, this can make for a great deal of confusion. But it might also offer a richer, truer sense of the world than the one perceived by those of us who have never been blind.
Sitting in the lab one day, MacLeod, smirking like a schoolboy who's hatching a prank, slides a drawing across the table to May. On the paper are four cubes. The top right cube and the bottom left cube are dark; the other two are light. The drawing is shaded as if light were coming from above, so the tops of the squares are lighter than their fronts. This makes the top of the dark square the same shade as the front of the light square. Experience tells us that the top of the dark cube has been brightened by a hidden light, but it still seems darker than the front of a light cube. It's an illusion based on knowledge. Naturally, May doesn't fall for it.
"He's actually closer to reality," Fine says. "We once showed him two circles— a small one close to him and a larger one farther away. To you or me they would have appeared to be the same size. But when we asked, 'What's the apparent size?' he couldn't understand. He kept saying, 'I know it's bigger because it's far away.'" Similarly, May's tactile experience with hallways and highways tells him that their sides are parallel, so he simply can't perceive converging lines of perspective. "A hallway doesn't look like it closes in at all," he says. "I see the lines on either side of the path, but I don't really think of them as coming closer in the distance." He pauses to mull this over. "Or maybe my mind doesn't believe what my mind is perceiving. When I see an object, it doesn't look different to me as I circle around it. I know orange cones around vehicles are cones because of context, not because I'm seeing the shape. If I picture looking down on a cone, it still looks like a cone."
Learning to see, for May, is really about learning to fall for the same illusions we all do, to call a certain mass of colors and lines his son, to call another group of them a ball.
One April morning, only weeks after his eye surgery, May took his skis and his family up to the Kirkwood Mountain Resort in the Sierra Nevadas— a place he knew like the texture on the back of his hand. This was where he had first learned to ski and where he had later met his wife. The sun was out, the trees were green (greener than he'd imagined), and the slopes were surrounded by gorgeous cliffs (were they miles away or just a few hundred yards?). As the lift churned above, skiers in puffy parkas flitted by, popping into his field of vision. His wife, acting as a guide, had to remind him to stop gawking and ski.
With only one working eye, May already lacked depth perception. But he also had little experience reading the shades and contours of a landscape. Heading down the mountain, he could hardly distinguish shadows from people, poles, or rocks. At first, he tried to compute the lay of the land consciously: If a certain slope was being lit from the side and a shadow fell in such a way, then the slope must be convex. But once he hit his first bump, he was tempted to close his eyes and ski the way he knew and loved.
Only a handful of adults have ever seen the world through the eyes of a newborn, and many who did came away wishing they were still blind. Their family and friends had convinced them that vision would offer a miraculous new appreciation and understanding of the world. Instead, even the simplest actions— walking down stairs, crossing the street— became terrifyingly difficult. Dispirited and depressed, about a third of them reverted back to the world of the blind, preferring dark rooms and walking with their eyes shut.
If May feels differently, it may be because his expectations were so low. For a man who used to enjoy windsurfing blind and alone, able more often than not to return to the pier from which he'd started, sight is just another adventure in a life of invigorating obstacles. Two years after his return to KirkwoodMountain, May has learned to match what he sees on a ski slope with his repeated physical experience of it. "He has jury-rigged himself quite a functional little system," Fine says. "He knows that this kind of shadow makes this bump, this kind makes another." Instead of closing his eyes on even the easiest slopes, he can now negotiate moguls without a guide.
"People have this idea that it's so overwhelmingly practical to have sight," May says. "I say it's great from an entertainment point of view. I'm constantly looking for things that are unique to vision. Running and catching a ball is one of them— I've been chasing balls my whole life. Seeing the difference between the blue of my two sons' eyes is another. Or if you drop something, you can find it."
The gift of sight may seem most miraculous, in the end, to those who have never been blind. But May still finds things in the world to entrance him. Sitting in the passenger seat of Fine's car one day, with his dog, Josh, panting at his feet, he ignores the blue Pacific to the left, the towering, top-heavy eucalyptus trees lining the road like something out of Dr. Seuss. Instead, he gazes at the beam of sunlight filtering through the window onto his lap. "I can't believe the dust is just floating in the air like this," he says. Oceans and trees, Seussean or otherwise, he has known all his life through touch. But this glitter of dust, suspended in the bright La Jolla sun, is an entirely new awareness. He waves his hand through the sparkling beam. "It's like having little stars all around you."